BR112012002082B1 - VEHICLE DRIVING SUPPORT DEVICE AND VEHICLE DRIVING SUPPORT PROCESS - Google Patents

Abstract

vehicle steering support device and vehicle steering support process when a lateral position in the vehicle's track width (x2obst + x0) direction (mm) reaches a predetermined control start position (60) being a position Track Width Steering Side (x2obst + x0) serving as a vehicle approach prevention indicator (mm), a control start is determined and a deviation moment (ms) in the center direction of a travel range (200) is applied to the vehicle (mm) to control the vehicle (mm). then, when the lateral position of the track width direction (x2obst + x0) of vehicle mm moves from outside to the control start position (60), the control start determination is suppressed for a period of predetermined time compared to a period prior to movement into the control start position (60).

Description

“VEHICLE STEERING SUPPORT DEVICE and VEHICLE STEERING SUPPORT PROCESS”

Technical Field

The present invention relates to a vehicle steering support and a vehicle steering process that supports a driver's steering through vehicle control so that the vehicle moves towards the center of a lane when in a lateral position of the vehicle reaches a predetermined lateral position in a lane width direction.

Background of the Technique

A technique described in patent literature 1 is an example of a conventional vehicle steering support device. In this conventional technique, when a vehicle's lane leakage trend is detected based on a lane dividing line and a future position of the vehicle after a predetermined period of time, the vehicle is controlled to move in a direction such that lane departure is avoided.

Citation List

Patent Literature 1: Publication of Japanese Unexamined Patent No. 2000-33860 ·

Summary of the Invention

However, in a system that causes an exit avoidance control to intervene when the vehicle's future position (future side position) or current side position exceeds a predetermined reference value, the control can intervene again in the state where the vehicle is traveling in one direction to cancel the lane departure trend.

In the case where the future vehicle position, in particular, is predicted, the expected future vehicle position tends to be unstable due to the driver's steering.

Thus, when carrying out a control intervention it is determined based on the position of the future vehicle as in the technique described in patent literature 1 mentioned before Petition 870170030306, of 05/08/2017, p. 9/67

2/54 previously, a new control that creates a sensation of strangeness is likely to be performed.

In this sense, an object of the invention is to provide a vehicle steering support device and a vehicle steering support process that are capable of properly carrying out a vehicle steering support control against a side obstacle while suppressing the feeling of strangeness given to the driver.

To solve the problems described above, a vehicle drive support device of a first aspect of the invention is characterized by comprising: a control start determination part configured to determine a control start when a lateral position of a vehicle in a bandwidth direction it reaches a control start position with a predetermined lateral position in the bandwidth direction; a vehicle controller configured to control the vehicle by applying a detour moment towards the center of a vehicle travel / traffic lane when the control start determination part performs the control start determination; and a control suppression part sets to, when the vehicle's lateral position in the lane direction moves from an out of control start position in the lane direction to a position approaching the travel range of 20 vehicle within the start control position in the direction of the lane, suppresses the control of applying moment of deviation to the vehicle for a predetermined period after the lateral position of the vehicle in the direction of the lane moves to the position of vehicle travel range approximation, compared to a period before moving to the position approaching the vehicle travel range.

In addition, a vehicle steering support process of a second aspect of the invention is characterized by comprising: a control start determination step determination of control start when a side position of a vehicle in a lane width direction reaches a starting position of controPetição 870170030306, of 05/08/2017, p. 10/67

3/54 l and a predetermined lateral position in the direction of bandwidth; a vehicle control vehicle control step by applying a detour moment towards the center of a vehicle travel lane for the vehicle; and a control suppression step of when the vehicle's lateral position in the lane direction moves from an out of control start position in the lane direction to a position approaching the vehicle travel lane within the control start position in the direction of the lane, suppressing the control of application of the moment of deviation for the vehicle for a predetermined period of time after the lateral position of the vehicle in the direction of 10 lane width moves to the position approaching the vehicle travel range, compared to a period before moving to the position approaching the vehicle travel range.

In the invention, when the vehicle's lateral position in the bandwidth direction moves to a position within the control start position after the control start 15 is determined, the control is suppressed for the predetermined period after the movement to the position over the interior. Thus, even when the vehicle's lateral position in the lane direction is unstable, control can be suppressed. As a result, a sense of awkwardness provided to the driver can be reduced.

Brief Description of Drawings

Fig. 1 is a schematic configuration diagram of a vehicle steering support device of embodiments based on the invention.

Fig. 2 is a block diagram schematically showing the processing of a brake / steering force control unit.

Fig. 3 is a flowchart showing a procedure for processing the brake / steering force control unit in a first run.

Fig. 4 is a schematic view showing a relationship between having a vehicle and an obstacle.

Petition 870170030306, of 05/08/2017, p. 11/67

4/54

Fig. 5 is a flowchart showing a first example of a start control position fixing processing procedure.

Fig. 6 is a first example of a correction quantity calculation map.

Fig. 7 is a second example of a correction quantity calculation map.

Fig. 8 is a third example of a correction quantity calculation map.

Fig. 9 is a fourth example of a color quantity calculation map10.

Fig. 10 is a flowchart showing a second example of the start control position fixing processing procedure.

Fig. 11 is a graph showing the characteristics of a K2recv gain.

Fig. 12 is a view to explain an operation of the first embodiment of the invention.

Fig. 13 is a view for explaining an operation of a conventional technique.

Fig. 14 is a flowchart showing a procedure for processing a brake / steering force control unit in a second embodiment of the invention.

Description of Achievements

Embodiments of the invention are described below based on the drawings.

Achievements are described for cases where a vehicle steering support device is installed in a rear wheel steering vehicle. Note that a vehicle in which the vehicle steering support device is installed may be a front wheel steering vehicle or a four wheel steering vehicle.

(First Realization) (Configuration)

Petition 870170030306, of 05/08/2017, p. 12/67

5/54

Fig. 1 is a schematic configuration diagram of a first embodiment device.

Reference numeral 1 in the drawing is a brake pedal. The brake pedal 1 is connected to a master cylinder 3 via a reinforcer 2. In addition, numeral 5 reference 4 in the drawing is a reservoir.

Master cylinder 3 is connected to wheel cylinders 6FL, 6FR, 6RF, 6RR of the respective wheels via a fluid pressure circuit 30. Thus, in a state where a brake control is not operated, master cylinder 3 increases the pressure of brake fluid according to an amount by which a driver steps on the brake pedal. The increased brake fluid pressure is supplied to the 6FL, 6FR, 6RF, 6RR wheel cylinders of the respective 5FL, 5FR, 5RF, 5RR wheels via the fluid pressure circuit 30.

A brake fluid pressure control unit 7 controls an actuator 30A in the fluid pressure circuit 30, and controls brake fluid pressures for the respective 15 wheels 5FL, 5FR, 5RF, 5RR, individually. In addition, the brake fluid pressure control unit 7 controls the brake fluid pressures of the respective wheels 5FL, 5FR, 5RF, 5RR in such a way that brake fluid pressure values correspond to command values sent from a brake / steering force control unit. A proportional solenoid valve that can control 20 hydraulic pressures from the respective 6FL, 6FR, 6RF, 6RR wheel cylinders to be the desired hydraulic brake pressures can be used as the 30A actuator.

Here, for example, a brake fluid pressure control unit used in anti-slip control (ABS), traction control (TCS), or vehicle dynamics control system (VDC) can be used as the unit of 25 brake fluid pressure control 7 and 30 fluid pressure circuit. The brake fluid pressure control unit 7 can be configured to control the brake fluid pressures of the respective 6FL, 6FR, 6RF wheel cylinders , 6RR, by itself only, that is, without the fluid pressure circuit 30. When the control unit

Petition 870170030306, of 05/08/2017, p. 13/67

6/54 brake fluid pressure 7receives the brake fluid pressure command values from the brake force / steering control unit 8 to be described later, the brake fluid pressure control unit 7 controls the pressures brake fluid according to the brake fluid pressure command values.

In addition, the vehicle is provided with a steering torque control unit 12.

The steering torque control unit 12 controls steering torques of the respective rear wheels 5RL, 5RR being steering wheels. This control is achieved by controlling a steering state of an engine 9, a selection of the gear ratio of an automatic transmission 10, and a throttle opening degree of an throttle valve 11. In other words, the unit steering torque control 12 controls an amount of fuel injection and ignition timing. In addition, the throttle opening degree is controlled at the same time. Thus, the steering state of motor 9 is controlled.

In addition, the steering torque control unit 12 produces steering torque values Tw being information used to control the brake / steering force control unit 8.

Note that the steering torque control unit 12 can control the Tw steering torques of the respective rear wheels 5RL, 5RR, only on its own, that is, without the brake / steering force control unit 8. However , when steering torque control unit 12 receives steering torque control values from brake / steering force control unit 8, steering torque control unit 12 controls steering torque Tw according to the steering torque command values received.

In addition, an image capture unit 13 with an image processing function is provided on a front portion of the vehicle. Image capture unit 13 is used to detect the position of an MM vehicle on a travel strip (see Fig. 4). Image capture unit 13 includes a camera

Petition 870170030306, of 05/08/2017, p. 14/67

7/54 monocular formed, for example, a CCD camera (load coupled device).

Image capture unit 13 captures a front image of the vehicle

MM. Then, the image capture unit 13 performs image processing on the captured frontal image of the MM vehicle, detects a line dividing the track 5 such as a white line 200 (track marker) (see Fig. 4), and detects the travel range at the bases of the detected white line 200. Next, the image capture unit 13 detects the travel range width of a vehicle travel range.

In addition, the image capture unit 13 determines a certainty of white line recognition to be described later.

In addition, the image capture unit 13 calculates a frontal angle (angle of deviation) ^ between the travel range of the MM vehicle and the front rear steering axis of the MM vehicle, an Xfrontal lateral offset with respect to the travel range, a Pfrontal travel strip curvature, and the like at the bases of the detected travel strip. Image capture unit 13 produces the calculated offset angle

Çtrontal, Xfrontal lateral displacement, Pfrontal travel range curvature, and the like for the brake / steering force control unit.

The image capture unit 13 detects the white line 200 forming the travel range, and calculates the angle of forward deviation at the bases of the detected white line. Thus, the detection accuracy of the front offset angle is affected by the detection accuracy 20 of the white line 200 of the image capture unit 13.

Note that the curvature of the Pírontal travel strip can be calculated based on the δ steering angle of a steering wheel 21 to be described later.

In addition, the vehicle includes 24L / 24R radar devices. The 24L / 24R radar devices are each a sensor for detecting an SM lateral obstacle (See

Fig. 4) on the right or left of the MM vehicle. Each of the radar devices

24L / 24R is, for example, a milli-wave radar capable of detecting the existence of the SM obstacle in a predetermined blind area on the sides and rear of the vehicle

Petition 870170030306, of 05/08/2017, p. 15/67

8/54 by emitting an electromagnetic wave to a K-AREA obstacle detection area being the predetermined blind area and by detecting a reflected wave from the emitted electromagnetic wave. In the following, 24L / 24R radar devices are also simply referred to as milli-wave radars. It is assumed that the 24L / 24R radar devices are capable of detecting a POSXobst relative side position, a DISTobst relative longitudinal position, and a dDISTobst relative longitudinal speed, which are relative to the SM obstacle, on each of the left and right sides. In addition, the 24L / 24R radar devices determine an obstacle recognition certainty to be described later.

Note that, the lateral direction in the realizations of the description refer to a direction of width of the strip, and the longitudinal direction refers to a direction extending from the strip.

In addition, the vehicle includes a master cylinder pressure sensor 17, an accelerator gap degree sensor 18, a steering angle sensor 15 19, an indicator switch 20, and wheel speed sensors 22FL, 22FR,

22RL, 22RR.

The master cylinder pressure sensor 17 detects an outlet pressure from the master cylinder 3, that is, a hydraulic pressure from the master cylinder Pm. The throttle opening degree sensor 18 detects the amount by which the throttle pedal is actuated, i.e., a throttle opening degree 0t. The wheel steering sensor 19 detects the steering angle δ of steering wheel 21. The indicator switch 20 detects an indicator operation of an indicator. The wheel speed sensors 22FL, 22FR, 22RL, 22RR detect rotation speeds of the respective wheels 5FL, 5FR, 5RF, 5RR, or so-called wheel speeds Vwi (i = fl, fr, rl, rr). These sensors and the like send the detected detection signals to the brake / steering force control unit 8.

Fig. 2 is a block diagram schematically showing the processing of the brake / steering force control unit 8. The processing of the uniPetition 870170030306, from 05/08/2017, p. 16/67

9/54 brake / steering force control 8 is performed based on a flow chart (Fig. 4) to be described later. In Fig. 2, processing is described schematically as blocks.

As shown in Fig. 2, the brake / steering force control unit 8 includes a future position predictor 8A, an avoidance control start detector (control start determination part) 8B, and a vehicle controller 8C. The 8B avoidance start control detector includes a start timing adjuster (control override part) 8Ba.

The future position predictor 8A predicts a future vehicle position (a future vehicle position in the direction of travel width) after a forward observation time period Tt being a pre-fixed fixation period, in the bases of a steering input for the driver, the status of the MM vehicle, and the like detected by the sensors and the image capture unit 13. Note that, a process of forecasting the future position of the vehicle is described later15.

The avoidance control start detector 8B detects a control start when the future vehicle position reaches a predetermined control start position 60 (a predetermined lateral position in the direction of bandwidth, see Fig. 4 to be described later ) in a case where an obstacle protection unit 25 is judged to be detecting the SM obstacle on either side of the vehicle.

In addition, when a future vehicle position 150 moves from an out of control start position 60 in the direction of bandwidth to a position approaching the travel range within control start position 60 in the direction of width range, the 8Ba start timing adjuster suppresses (decreases the frequency of the control start determination or reduces a control amount) the control (the control start determination and the control amount) up to a retention time period control status when a predetermined period of time elapses after the future vehicle position.

Petition 870170030306, of 05/08/2017, p. 17/67

10/54 approaching the vehicle travel range, compared to a period before moving to the position approaching the vehicle travel range.

When the avoidance control start detector 8B detects the control start, vehicle controller 8C calculates a deviation moment Ms used to control the MM vehicle in such a way that the MM vehicle is avoided from approaching the FM obstacle. The deviation moment Ms is one such deviation moment that is used to control the MM vehicle towards the center of the lane.

Note that the obstacle detection unit 25 of Fig. 2 corresponds to the 24L / 24R radar devices, and detects information about the obstacle SM where the MM vehicle 10 is used as a reference. The information includes the existence of the SM obstacle in the K-AREA obstacle detection area for the sides and rear of the vehicle, as well as the relative lateral position POSXobst, the relative longitudinal position DISTobst, a relative longitudinal speed dDISTobst, and the like of the SM obstacle with respect to the MM vehicle.

Fig. 3 is a flowchart showing an obstacle avoidance control processing procedure performed by the brake / steering force control unit 8.

Obstacle avoidance control processing is performed by performing a scheduled time interruption at each predetermined sample time period (control cycle) AT (for example, every 10 ms). Note that the process shown in Fig. 3 does not include a communication process. However, information acquired from a calculation process is updated and stored in a storage device when necessary, and necessary information is read from the storage device when necessary.

<Step S10>

First, in step S10, the brake / steering force control unit 8 reads various types of data from the sensors, controller, and control units described above.

Specifically, the brake / steering force control unit 8 acquires

Petition 870170030306, of 05/08/2017, p. 18/67

11/54 the wheel speeds Vwi (i = fl, fr, rl, rr), the steering angle δ, the degree of throttle opening 0t, and the hydraulic pressure of master cylinder Pm detected by the wheel speed sensors respectively 22FL, 22FR, 22RL, 22RR, the steering angle sensor 19, and throttle opening degree sensor 18, and the master cylinder pressure sensor 17. In addition, the brake / steering force control unit 8 acquires the indicator switching signal from the indicator switch 20, the φfront offset angle, the Xfrontal side offset, and the Pfrontal travel range curvature detected by the image capture unit 13, and the SM side obstacle information detected by the devices of 24L / 24R radar (10 obstacle detection unit 25).

<Step S20>

Then, in Step S20, the brake / steering force control unit 8 calculates vehicle speed V. Specifically, vehicle speed V is calculated from the following formulas on the basis of the wheel speeds Vwi 15 detected by the sensors wheel speed 22FL, 22FR, 22RL, 22RR.

V = (Vwrl + Vwrr) / 2 (in the case of front wheel steering),

V = (Vwfl + Vwfr) / 2 (in the case of rear wheel steering) ...... (1) where Vwfl, Vwfr are the wheel speeds of the front left and right wheels respectively, and Vwrl, Vwrr are respectively the wheel speeds of the 20 left and right rear wheels. In other words, in the formula (1) mentioned above, the vehicle speed V is calculated as an average value of the wheel speeds of the non-steering wheels. Note that, since the vehicle in the first realization in a rear wheel steering vehicle, vehicle speed V is calculated using the last formula, that is, the wheel speeds Vwfl, 25 Vwfr, of the front left wheels and right 5FL, 5FR.

In addition, when a different automatic brake control device such as an ABS control (anti-lock brake system) is operating, an estimated vehicle speed estimated by the different brake control device is acquired, and

Petition 870170030306, of 05/08/2017, p. 19/67

12/54 is used as vehicle speed V.

<Step S30>

In step S30, the brake / steering force control unit 8 acquires Lobst-Robst stocks of the SM obstacle in the left and right areas of the MM vehicle, 5 on the bases of the 24L, 24R radar device signals respectively on the left side and right. In addition, the brake / steering force control unit 8 acquires the position and speed of the SM side obstacle in relation to the MM vehicle. Here, as shown in Fig. 4, the areas to the side of the MM vehicle include obliquely rear areas of the MM vehicle.

The K-AREA obstacle detection area shown in Fig. 4 is fixed in predetermined longitudinal and lateral positions in the area to the side of the MM vehicle. In addition, the longitudinal position can be fixed so that the higher the speed of the SM obstacle in relation to the MM vehicle is, the larger the K-AREA obstacle detection area is.

<Step S40>

Then, in step S40, the brake / steering force control unit 8 reads the Xfrontal lateral (lateral position) displacement of the MM vehicle on a travel road on which the MM vehicle is currently traveling and the curving of the lane Pfrontal trip, from the image capture unit 13.

Note that the curvature of the Pírontal travel strip can be acquired not only from a calculation based on the image captured by the image capture unit 13, but also in the way that follows. For example, travel lane curvature information at a vehicle position can be acquired based on map information stored in a navigation system.

In addition, the frontal deviation angle ^ of the MM vehicle with respect to the travel road on which the MM vehicle is currently traveling is calculated. The forward offset angle ^ is used to detect a travel state in the lane.

In the first embodiment, the angle of frontal deviation ^ is detected by, for example,

Petition 870170030306, of 05/08/2017, p. 20/67

13/54 example, conversion of frontal vehicle image captured by image capture unit 13 into a bird's eye view image and using an angle of the white line 200 (lane marker) with respect to an up-down direction of converted image.

Note that the angle of frontal deviation ^ can be calculated based on the white line 200 next to the MM vehicle in the image captured by the image capture unit 13. In this case, the angle of frontal ^ deviation is calculated from formula (2) described below by using, for example, a change amount of the Xfrontal lateral displacement of the MM vehicle.

$ frontal = tan ^-1 (dX '/ V (= dX / dY)) ........ (2) where dX is a change quantity of the Xfrontal lateral displacement per unit time, dY is a change quantity in the direction of travel per unit of time, and dX 'is a derivative of the amount of change dX.

Note that, in the case of calculating the angle of atrial deviation at the bases of the next white line 15, the calculation is not limited to one where the angle of the frontal deviation is calculated using the Xfrontal lateral displacement as in the formula (2) mentioned above. For example, the calculation can be done as follows. The white line 200 detected nearby is extended to a distant position, and the Angle of Forward deviation is calculated based on the extended white line 200. These calculation processes of Xfrontal lateral displacement, Pfrontal travel range curvature, the Angle of Forward deviation , and similar to the MM vehicle on the bases of the vehicle front image, publicly known techniques already used in various devices that control the MM vehicle through line recognition 200, such as a travel control device keeping track, and so on. described in detail.

<Step S50>

In step S50, the brake / steering force control unit 8 calculates a neutral offset rate AAminho on the bases of formula (3) described below. The Ermine neutral deviation rate is a required deviation rate to make the MM vehicle

Petition 870170030306, of 05/08/2017, p. 21/67

14/54 keep traveling along the travel road. When the MM vehicle is traveling on a straight road, the neutral deviation rate φ ^^ is zero. However, the neutral deviation rate φ ^^ changes on a curved road due to the forward curvature of the curved road. Thus, the Pfrontal travel range curvature described above is used to calculate the neutral taxa path deviation rate.

φ path = Front. V .......... (3)

Here, the neutral deviation rate φ path required for the MM vehicle to keep traveling along a travel path can be calculated in a simplified way using an average value em ^^ over a predetermined period or a value obtained by multiplying the neutral deviation rate φ ^^ through a filter with a large time constant.

<Step S60>

In step S60, the brake / steering force control unit 8 sets the forward observation time period Tt. Specifically, a predetermined forward observation time period Tt0 is fixed as the forward observation time period Tt as in the following formula.

Tt ^ Tt0

The forward observation time period Tt10 is a time used to predict a situation where the driver causes the MM vehicle to approach the MM obstacle in the future. For example, the Tt0 forward observation time period is fixed to one second.

Next, a alvomΘtΘrista target deviation rate and a driver ψcΘrrection corrected target deviation rate are calculated.

The target deviation rate is calculated from the δ steering angle and the vehicle speed V as shown in the formula described below. The target deviation rate ψι ^ η ^ is a deviation rate that the driver tries to generate by performing a steering operation. In other words, the target ψmΘtΘrista deviation rate is a deviation rate that the driver deliberately tries to generate.

Petition 870170030306, of 05/08/2017, p. 22/67

15/54 ψιτιοΐοΓίβΐα = Kv. δ. V (4) where Kv is a gain setting according to the vehicle's specifications and the like.

In addition, the corrected target deviation rate de ^ ί ^ of driver is calculated from the formula described below. The driver-corrected target deviation rate is a value obtained by subtracting the neutral taxa'path deviation rate required for the MM vehicle to keep traveling along the travel road from the ψmotoΓist target deviation rate. Thus, a steering effect performed to make the MM vehicle travel on the curved road is removed from the driver's target deviation rate.

ψcoΓΓection of driver = ψmotorist ^_ φ path ........ (5)

In other words, the target deviation rate corrected ψcoΓΓection of driver is one of deviation between the deviation rate (neutral deviation rate φ ^^) required for the MM vehicle to keep traveling along the travel range and the deviation rate (target deviation rate ψmotorist) that the driver tries to generate by performing a steering operation, and is a deviation rate corresponding to the driver's intention to change lanes <Step S70>

Then, in step S70, the brake / steering force control unit 8 calculates an expected position of vehicle AXb on the basis of formula (6) described below, the expected position of vehicle AXb being a distance from a lateral position current (position in the direction of travel road width) of the MM vehicle to a lateral position of the MM vehicle after the forward observation time period Tt. In other words, the lateral distance (distance in the traveling road width direction) from the current lateral position of the MM vehicle to the MM vehicle position after the forward observation time period Tt (future vehicle position 150) be calculated as the expected position of vehicle AXb. Here, the forward observation time period Tt is a value fixed with the appropriate, and is a design value. Note that the predicted vehicle position AXb is used in determining

Petition 870170030306, of 05/08/2017, p. 23/67

16/54 if the avoidance control against the SM obstacle starts.

AXb = (K1 ^ frontal + K2Çm + K3Çm ’) ...... (6) where ^ frontal = angle of deviation

Çm; target deviation angular speed

Çm ’: angular acceleration of target deviation

The angular velocity of target deviation mentioned above Çm is expressed by the following formula.

Çm = ψcorrection of driver. TT (7)

In addition, the angular acceleration of target deviation Çm 'is expressed by the following formula.

Çm '= Çm. TT ² (8)

Here, when an observation distance L is used to obtain and predicted vehicle position AXb is a dimension of the angle of deviation, the predicted vehicle position AXb can be expressed by the following formula.

AXb = L. (K1 Çfrontal + K2Çm.Tt + K3 Çm'.Tt ² ) (9)

Here, a relationship between the forward observation distance L and the forward observation time period Tt is expressed by the following formula.

forward observation distance L = forward observation time period Tt. vehicle speed V (10)

In view of these characteristics, a clamping gain K1 is a value using vehicle speed V as a function. In addition, a fixation gain K2 is a value using vehicle speed V and the forward observation time period Tt as functions. A fixation gain K3 is a value using vehicle speed V and the square of the forward observation time period Tt as functions.

Note that the predicted position of the MM vehicle can be calculated by obtaining the steering angle component and the steering speed component individually and then performing a high selection, as in the formula

Petition 870170030306, of 05/08/2017, p. 24/67

17/54 follows.

, - \ Xb = Max (K2Çm, K3fom ') ....... (11) <Step S80>

Then, in step S80, the brake / steering force control unit 8 sets a determination limit used to determine whether to initiate control or not. This determination limit is a determination limit used to determine whether to start avoidance control against the SM obstacle, and corresponds to the control start position 60 described above.

In the first realization, the determination limit is fixed based on an obstacle distance X2obst being a predetermined distance shown in Fig.

4. The obstacle distance X2obst corresponds to a distance from a predetermined virtual position (position in the direction of bandwidth) where the obstacle SM exists for the white line 200 in the lateral direction.

Here, an XY coordinate system is used where a Y axis extends in a direction parallel to the travel road (longitudinal direction) and an X axis extends in a direction orthogonal to the travel road, that is, a direction of bandwidth (lateral direction). Then, the obstacle distance X2obst is fixed on the coordinate of the X axis. Note that the obstacle distance X2obst is zero when the virtual predetermined position where the obstacle exists in a position on the white line 20, is a positive value when the virtual predetermined position is outside the white line 200, and is a negative value when the virtual predetermined position is inside the white line 200.

As described above, a value obtained by adding a predetermined obstacle distance X2obst for a lateral displacement X0 of the MM vehicle in Fig. 4 (ie, X2obst + X0) is used as a virtual distance from the MM vehicle for the obstacle SM, and thus the determination limit is fixed. Note that, X0 is the lateral steering distance (distance in the width direction) between the MM vehicle and the white line 200 as in the illustration in Fig. 4. The lateral displacement X0 is

Petition 870170030306, of 05/08/2017, p. 25/67

18/54 acquired, for example, by processing an image captured by the image capture unit 13. The distance in the lateral direction (lateral displacement X0) between the MM vehicle and the white line 200 is a positive value when the MM vehicle is within the white line 200, and is a negative value when the MM vehicle is in a position 5 beyond the white line 200. Note that the X0 lateral displacement of the MM vehicle is obtained using the Xfrontal lateral displacement detected by the image capture unit. image 13.

In the following, an X2obst obstacle distance setting procedure is described.

Fig. 5 is a flowchart showing a first example of a start control position processing procedure in step S80.

<Step S801>

First, in step S801, the brake / steering force control unit 8 judges whether a Fout_obst obstacle avoidance control flag to be described later is switched off or not. When Fout_obst = OFF, the start control position setting processing proceeds to step S802. When Fout_obst = on, the start control position fixing processing is immediately completed.

Note that, although described in detail later, the Fout_obst obstacle avoidance control flag is basically a flag which is switched on when the predicted vehicle position AXb described above reaches or exceeds X2obst + X0. In other words, the Fout_obst obstacle avoidance control flag is a flag that is set to on when the side position of the vehicle after the forward observation time period Tt coincides with a position of the obstacle distance X2obst from of the white line 200 25 or is in a position farther from the range than the position of the obstacle distance X2obst.

<Step S802>

In step S802, the brake / steering force control unit 8 judges whether

Petition 870170030306, of 05/08/2017, p. 26/67

19/54 a previous value of the obstacle avoidance control determination flag Fout_obst is on or not. When Fout_obst (previous value) = on, processing proceeds to step S803. When Fout_obst (previous value) = off, processing proceeds to step S807 to be described later.

<Step 803>

In step S803, the brake / steering force control unit 8 sets a Freturn return flag to “1”, and processing proceeds to step S804. The Freturn return flag is a flag that is set to “1” when it is judged in step s801 that the current obstacle avoidance control determination flag 10 Fout_obst + off and it is judged in step S802 that the previous flag determination obstacle avoidance control Fout_obst = on. In other words, the Freturn return flag is a flag indicating whether the predicted vehicle position AXb has become (returned to) AXb <X2obst + X0 after reaching or exceeding the determination limit X2obst + Xo (obstacle distance X2obst + displacement 15 lateral X0).

<Step S804>

In step S804, the brake force / steering control unit 8 sets a correction amount δ used to set a corrected obstacle distance X2obst_h to be described later. Here, the correction amount δyc can be a fixed value or it can be fixed while referring to a predetermined correction amount calculation map as described later.

As shown in figure 6 (a first example of the correction quantity calculation map), the correction calculation quantity map has a vertical axis representing the correction quantity δyc and a horizontal axis representing a bandwidth. Here, a vehicle travel lane bandwidth can simply be used as the lane width, or a distance between the MM vehicle and a white line on the side opposite the side on which the SM obstacle is the control target. avoidance exists can be used with the bandwidth. Beyond

Petition 870170030306, of 05/08/2017, p. 27/67

In addition, the smaller the bandwidth, the greater the amount of correction õyc is fixed.

Note that, in Fig. 6, if there is an obstacle on the opposite side (obstacle on the opposite side on which the MM vehicle is to move laterally through obstacle avoidance control), the amount of correction õyc it can be fixed according to a distance between the MMM vehicle and the obstacle on the opposite side instead of the lane width. In this case, the shorter the distance between the MM vehicle and the obstacle on the opposite side is, the greater the fixing amount of correction õyc.

In addition, when the control start position 60 is fixed based on the position of the white line 200 (when the limit of determination is fixed based on

X2obst), the correction amount õyc can be calculated based on a recognition status of the white line 200. In this case, as shown in Fig. 7 (a second example of the correction quantity calculation map), a map is used that it has a vertical axis representing the amount of correction õyc and a horizontal axis representing the certainty of recognition of white line. The certainty of white line recognition (recognition accuracy) is detected, for example, of an amount of a high frequency component of the white line 200. In this case, the greater the noise and deflection components, or the further the edge of the white line 200 is blurred (the certainty of white line recognition is 20 lower), the greater the amount of correction õyc is fixed.

In addition, the control start position 60 can be fixed based on a distance to the side obstacle SM. In this case (case where a relative distance in lateral direction AO between the vehicle MM and the lateral obstacle SM illustrated in Fig. 4 is used as the limit of determination), the correction amount õyc is calculated 25 based on a state of recognition of the SM obstacle. This calculation of the correction amount õyc is performed as follows. For example, a map having a vertical axis representing the amount of correction õyc and a horizontal axis representing the certainty of obstacle recognition as shown in Fig. 8 (a

Petition 870170030306, of 05/08/2017, p. 28/67

21/54 third example of the correction amount calculation map) is fixed in advance, and this map is used to calculate the correction amount õyc based on the certainty of obstacle recognition. Here, the certainty of obstacle recognition (recognition accuracy) is detected, for example, from a quantity 5 of a high frequency component of lateral position information of the SM obstacle detected by the radar. The greater the components of noise and deflection of the lateral steering distance (relative distance in the lateral direction AO) between the MM vehicle and the SM obstacle are, or the more the edge of the SM obstacle is blurred (the less certainty of obstacle recognition is) ), the greater the amount of correction õyc fixed10 da.

In addition, the correction amount õyc can be calculated based on a state of movement of the MM vehicle. In this case, as shown in Fig. 9 (a fourth example of the correction amount calculation map), a map having a vertical axis representing the correction amount õyc and a horizontal axis representing an amount of movement change of the MM vehicle is fixed in advance, and this map is used to calculate the amount of correction õyc based on the amount of movement change of the MM vehicle. Here, the amount of change in movement of the MM vehicle is a parameter that indicates the stabilization of the posture of the MM vehicle, and is, for example, amounts of change in an angular velocity of 20 steering, a rate of deviation, a lateral acceleration, and a lateral tire force per unit time which are detected respectively by the steering angle sensor 19, a deviation rate sensor, an acceleration sensor, and a lateral force sensor such as a vehicle posture detector. The greater the amounts of change (the more unstable the vehicle is), the greater the amount of correction 25 õyc is fixed. Note that the angular speed of steering, the rate of deviation, the lateral acceleration, and the lateral tire force can be used as they are instead of the amount of movement change. In such a case, the higher these values are, the greater the amount of correction õyc fixed.

Petition 870170030306, of 05/08/2017, p. 29/67

22/54

In addition, the amount of correction õyc can be fixed so that the longer the time lag after the obstacle avoidance control flag Fout_obst changes from on to off, the smaller the amount of correction õyc is fixed.

<SS805 step>

Then, in step S805, the brake force / steering control unit fixes the corrected obstacle distance X2obst_h by using the õyc correction amount set in step S804 mentioned earlier or step S812 to be described later. Here, the corrected obstacle distance X2obst_h is a value obtained by adding the correction amount õyc for a predetermined reference obstacle distance X2obst_0 (obstacle distance in a normal state) (X2obst_h + X2obst_0 + õyc).

<Step S806>

Next, processing proceeds to step S806 and the brake / steering force control unit 8 sets the obstacle distance X2obst on the basis of the corrected obstacle distance X2obst_h fixed in step S805 mentioned above and the predetermined reference obstacle distance X2obst_0 ( obstacle distance in a normal state). Here, the corrected obstacle distance X2obst_h and the reference obstacle distance X2obst_0 are compared with each other, and one with the highest value is fixed as the obstacle distance X2obst.

X2obst = Max (X2obst_h, X2obst_0) ....... (12)

Specifically, between a position whose distance from a white line 200 is the corrected obstacle distance X2obst_h and a position whose distance from the white line 200 is the reference obstacle distance X2obst_0, one that is furthest from the white line 200 is selected as the control start position 60.

Note that the obstacle distance X2obst can be limited by a predetermined value X2obst_limite. In this case, X2obst = Max (X2obst_h, X2obst_0, X2obst_ limit) is fixed.

Petition 870170030306, of 05/08/2017, p. 30/67

23/54

Note that, instead of the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0), the relative distance of lateral direction AO between the vehicle MM and the obstacle SM can be used as the determination limit for the start of control as described above (Fig. 4). The relative AO lateral steering distance is detected through the use of 24L / 24R radar devices (obstacle detection unit). In this case, as in the correction described above made for the obstacle distance X2obst, correction is carried out by adding correction value AO_h to the distance in relation to the lateral direction AO.

<Step S807>

In step S807, the brake / steering force control unit 8 judges whether the Freturn return flag is set to “1”. Then, when the Freturn return flag is “0” indicating that the return process is complete, processing proceeds to step S809 to be described later.

<Step S808>

In step S808, the brake force / steering control unit 8 sets the obstacle distance X2obst, and completes the start control position fixing processing. In this case, the predetermined reference obstacle distance X2obst_0 is fixed as the obstacle distance X2obst. In the first embodiment, the predetermined reference obstacle distance X2obst_0 is fixed to a predetermined distance to a position outside the white line 200. Specifically, the start control position 60 is fixed to a position that is outside the white line 200 and is away from the white line 200 by the predetermined distance. Note that the previously mentioned predetermined reference obstacle distance X2obst_0 can be set to 0 (that is, the control start position 60 is set to the white line position 200), or it can be set to a predetermined distance from the line white 200 to a position within the white line 200 (i.e., the control start position 60 is fixed to a predetermined lateral position within the white line 200). For example, the control start position 60 can be provided 870170030306, from 05/08/2017, p. 31/67

24/54 previously fixed according to the period of observation time forward Tt in the following way or similar. The forward observation time period Tt is fixed to 0 and the control start position 60 is fixed to a position that is within the white line 200 and away from the white line 200 by a predetermined distance.

<Step S809>

In addition, in step S809, the brake / steering force control unit 8 judges whether a predetermined period of time has elapsed since the return flag Freturn = 1 is set. Here, the predetermined time is the control state retention time period described above. If the predetermined period of time has not yet elapsed, processing proceeds to step S811.

Note that the control state retention time period can simply be the predetermined time period, but it can also be, for example, a period of time until a travel distance of the MM vehicle reaches a predetermined distance. In this case, the travel distance after the Freturn = 1 return flag is fixed is measured, and it is judged in step S809 if the measured travel distance has reached the predetermined distance. When the measured travel distance reaches the predetermined distance, processing proceeds to step S811. Alternatively, the holding period in the control state can be, for example, a period of time until the relative distance between the vehicle MM and the obstacle SM reaches or exceeds a predetermined distance after the return flag Freturn = 1 is set . As described above, the control state retention time period is a period of time that is appropriately changeable.

<Step S810>

Meanwhile, if the predetermined time (control state retention time) has elapsed after Freturn = 1 has been set, processing proceeds to step S810. Here, the Freturn return flag is set to the Freturn return flag = 0, and the processing then proceeds to step S808 mentioned abovePetition 870170030306, from 05/08/2017, pg. 32/67

25/54 mind.

<Step S811>

In step S811, the brake / steering force control unit 8 judges whether the lateral displacement X0 of the MM vehicle has increased, that is, whether the MM vehicle has moved towards the interior of the lane. This judgment can be made based on a deviation between a last value and a previous value (lateral displacement X0 in a previous calculation cycle) of the lateral displacement X0 of the MM vehicle, or based on a sign of the derivative of the lateral displacement X0 of the vehicle. MM. When the lateral displacement X0 increased, that is, the MM vehicle moved towards the interior of the strip, the processing proceeds to step S812. In other cases, the start control position fixing processing is immediately completed (X2obst is kept at a calculated value in the previous calculation cycle).

<Step S812>

In step S812, the brake force / steering control unit 8 performs a correction for the correction amount õyc fixed in the previous calculation cycle, and processing proceeds to step S805 mentioned above. Here, the correction amount õyc is corrected by calculating the deviation (referred to as lateral displacement deviation) between the last value and the previous value (lateral displacement X0 in the previous calculation cycle) of the lateral displacement X0 of the MM vehicle and then subtracting the lateral displacement deviation of the correction amount õyc fixed in the previous calculation cycle. Thus, when processing proceeds to step S805 via step S812, the corrected obstacle distance X2obst_h fixed in step S805 is decreased according to the movement (increase in lateral displacement X0) of the MM vehicle when the MM vehicle moves in the direction inside the track.

<Step S90>

Returning to Fig. 3, in step S90, the brake / steering force control unit 8 performs the control start determination.

Here, the brake / steering force control unit 8 determines the start of

Petition 870170030306, of 05/08/2017, p. 33/67

26/54 control when the following formula is satisfied in a state where SM lateral obstacle is detected.

AX2 = AXb-X0> X2obst ...... (13)

Specifically, as shown in Fig. 4, it is judged whether the lateral distance AX2 between the white line 200 and the future position 150 of the MM vehicle (position of the vehicle after the forward observation period Tt, and is also referred to as point observation point 150) reached or exceeded the obstacle distance X2obst. Here, the formula (13) mentioned above can be transformed as follows.

AXb> X2obst + X0 ....... (14)

In other words, it is judged whether the future vehicle position 150 is in the control start position 60 or is out of the control start position 60 in the direction of bandwidth, in the state where the side obstacle SM is detected.

Then, when the condition described above is satisfied, it is assumed that a lane change operation in the direction of SM or similar obstacle has been carried out and control over the SM obstacle is determined to be initiated. When control over the SM obstacle is determined to be initiated, the Fout_obst obstacle avoidance control flag is set on. In the meantime, if the condition described above is not satisfied, the Fout_obst obstacle avoidance control flag is set to off.

Note that when the relative distance in the AO lateral direction between the MM vehicle and the SM obstacle is used as the limit of determination for the control start, the control start is determined when the following formula is satisfied

AXb> AO ....... (15)

The predicted vehicle position AXb is actually obtained for each of the left and right side of the vehicle as AXbL / AXbR, and the determination is made individually for AXbL / AXbR.

In addition, the SM obstacle being the control target may include not only

Petition 870170030306, of 05/08/2017, p. 34/67

27/54 a vehicle to the rear and sides of the MM vehicle but also a vehicle approaching in an adjacent lane in front of the MM vehicle.

Here, when it is judged whether the future prediction position AXb is less than the limit of determination, a hysteresis of F can be provided in such a way that 5 AX2 <X2obst-F. In other words, a dead band can be provided. Specifically, the deadband can be provided between a control intervention limit and a control termination limit.

Also, Fout_obst is set to on when Fout_obst is off. In addition, a time condition such as elapsing from a predetermined period of time after Fout_obst is set to off can be added to the condition following Fout_obst to be set to on.

In addition, when a predetermined period of time T control elapses after determining the setting of Fout_obst to on, the Fout_obst + off can be fixed to complete the control. However, in the case where Fout_obst = des15 on is fixed after the predetermined period of time elapses from fixation determination of Fout_obst to on as described above, the lateral direction distance AX2 between white line 200 and future position 150 (point forward observation 150) of the MM vehicle is not necessarily less than the obstacle distance X2obst when Fout_obst = off is fixed. Likewise, in this case, processing of a flowchart of Fig. 10 is preferably performed rather than the processing illustrated in the flowchart of Fig.5 above.

The flowchart of Fig. 10 (a second example of the start control position fixing processing procedure) is provided with step S901 and step S902 processes in place of step S802 process in the flow diagram of Fig. 5 25 described above, and other processes are the same as those in the flow chart of Fig. 5. In the same way, the processes of step S901 and step S902 are described.

<Step S901>

In step S901, the brake / steering force control unit 8 judges whether the

Petition 870170030306, of 05/08/2017, p. 35/67

28/54 predicted vehicle position AXb is less than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0). In other words, the brake / steering force control unit 8 judges whether the future position 150 (forward observation point 150) of the MM vehicle is within the control start position 5l and 60 in the bandwidth direction. When the predicted position of vehicle AXb is less than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0), processing proceeds to step S902, and in other cases, processing proceeds to step S807 mentioned above. Note that, the determination limit in step S901 mentioned above may not be strictly X2obst + X0, and may include hys hysteresis in a way such as X2obst + X0-hys, for example.

<Step S902>

In step S902, the brake / steering force control unit 8 judges whether the predicted vehicle position AXb in the previous calculation cycle is equal to or greater than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0). In other words, the brake / steering force control unit 8 judges whether the MM vehicle's future position 150 (forward observation point 150) in the previous calculation cycle is the control start position 60, or is out of range. control start position 60 in the direction of bandwidth. When the vehicle position shown in AXb in the previous calculation cycle is equal to or greater than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0), processing proceeds to step S803 mentioned above, and in others cases, processing proceeds to step S807 mentioned above.

Here, the case where it is judged in step S901 that the predicted vehicle position 25 AXb is less than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0) (judged Yes in step S901) and is judged in step S902 that the predicted vehicle position AXb in the previous calculation cycle is equal to or greater than the determination limit X2obst + X0 (obstacle distance X2obst + displacement

Petition 870170030306, of 05/08/2017, p. 36/67

29/54 lateral X0) (judged Yes in step S902) is a case where the future position 150 (observation point forward 150) of the MM vehicle being out of control start position 60 in the direction of bandwidth in the cycle previous calculation moves to a position within control start position 60 in the direction of bandwidth in the last 5 calculation cycle.

Here, consider the case where the Fout_obst obstacle avoidance control flag is set to off even when the predicted vehicle position AXb is not less than the determination limit X2obst + X0 (obstacle distance X2obst + lateral offset X0) (for example, the case where the Fout_obst obstacle avoidance control flag is set to off after a predetermined period of time has elapsed since the Fout_obst obstacle avoidance control flag is set to on, the case where the Fout_obst obstacle avoidance control flag is set to OFF through the intervention of another control, or similar). In such a case, even when the Fout_obst obstacle avoidance control flag is deemed to be set to off in step S801, the predicted vehicle position AXb may not actually be less than the limit of determination X2obst + X0 (distance from obstacle X2obst + lateral displacement X0). Likewise, it is preferable that when the obstacle avoidance control determination flag 20 Fout_obst is judged to be off in step S801, the judgment processes of step S901 and step S902 are performed to judge whether future position 150 (point forward observation point 150) of the MM vehicle being out of control start position 60 in the bandwidth direction in the previous calculation cycle moves to a control start position 60 in the bandwidth direction in the pre25 calculation cycle vio moves to a position within control start position 60 in the direction of bandwidth in the last calculation cycle.

Note that, in the process of step S902, it can also be judged if the white line 200 is recognized in a previous calculation cycle. Specifically, it is judged

Petition 870170030306, of 05/08/2017, p. 37/67

30/54 in step S902 if the predicted vehicle position AXb in the previous calculation cycle is equal to or greater than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0), or if the white line 200 is recognized in the previous calculation cycle. When the predicted vehicle position AXb in the previous calculation cycle is 5 or greater than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0), or when the white line 200 is not recognized (or the accuracy recognition line 200 is low) in the previous calculation cycle, processing proceeds to process step S803. In other cases, processing proceeds to step process S807.

When the white line 200 is not recognized in the previous calculation cycle, the accuracy of the white line recognition is deteriorated, and thus there is a possibility of error in the lateral position of the recognized white line 200. Generally, the white line 200 is detected by capturing an image of a road surface towards the front of the vehicle, and then subjecting the captured image to edge processing 15 (processing for detecting luminescence shift points in the image).

However, the white line 200 may not be precisely detected on, for example, a snowy road, since the snow luminescence on the road surface is high. When the accuracy of recognition of the white line is deteriorated as in the case described above, or when there is an error in the recognized lateral position of the white line 200, a jolt (frequent change of the lateral position) occurs at the lateral position of the white line 200. This causes the value of the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement X0) to also change frequently. As a result, even when there is no change in the predicted vehicle position AXb, the predicted vehicle position AXb can often change from a value less than the determination limit X2obst + X0 (obstacle distance X2obst + lateral displacement

X0) for a value equal to or greater than that, or from a value equal to or greater than that for a value less than that. This can cause the control to become unstable.

Petition 870170030306, of 05/08/2017, p. 38/67

31/54

It is preferable that the following process is carried out in order to prevent the control from becoming unstable due to the accuracy of white line recognition. In step S902, it is judged whether the white line is recognized in the previous calculation cycle. When the white line is not recognized (or the white line recognition accuracy is low) in the previous calculation cycle, processing proceeds to the step processes S803, S804, S805, and the corrected obstacle distance X2obst_h is fixed.

In addition, while the obstacle avoidance control is performed, an execution direction control Dout_obst is judged from a judged direction of the predicted vehicle position AXb. When the predicted vehicle position AXb is on the left side, Dout_obst = left is fixed, and when the predicted vehicle position AXb is on the right side, Dout_obst = right is fixed.

Here, when anti-slip control (ABS), traction control (TCS), or vehicle dynamics control system (VDC) is operating, the Fout_obst obstacle avoidance control flag can be set to 15 off . This fixation is done, so that the obstacle avoidance control is not activated while the automatic brake control is activated.

<Step S100>

Then, in step S100, the brake / steering force control unit 8 performs a warning production process.

Here, the warning is produced when it is judged in the aforementioned step S90 that the start control position 60 (determination limit) is reached.

Note that, the warning can be produced before the forward observation point 150 (future position) based on the aforementioned forward observation time period Tt reaches the control start position 60. For example, the 25 time period of control forward observation Tt is multiplied by a predetermined Kbuzz gain (> 1) in such a way that the forward observation time period Tt is made longer than that used in the S90 step detection mentioned earlier. Then, the warning is produced when it is deemed that the observation point Petition 870170030306, of 05/08/2017, p. 39/67

32/54 forward section 150 (future position) calculated using (Tt-Kbuzz) in the bases of formula (6) mentioned above reached the control start position 60 in step S90 mentioned above.

In addition, it is possible for the notice to be produced when it is judged at the

S90 mentioned earlier that the operation of an obstacle avoidance system is to be initiated, and the control is initiated after the passage of a predetermined time from the production of warning.

In addition, it is possible that, in step S90, the determination of the control start is made using the determination limit set in step S80 mentioned 10 above, while in step 100, a pre-corrected determination limit (determination limit X2obst_0 in the normal state) is used to produce the warning. In this case, a warning start is not suppressed, and only the control start is actually suppressed.

<Step S110>

Then, in step S110, the brake / steering force control unit 8 sets a target deviation moment Ms.

When the Fout_obst obstacle avoidance control flag is off, the target deviation moment Ms is set to zero, and processing proceeds to step S120.

Meanwhile, when the Fout-obst obstacle avoidance control flag is set, the target deviation moment Ms is calculated from the following formula, and processing proceeds to step S120.

Ms = K1recv.K2recv.AXs ....... (16)

AXs = (K1mon. Çtrontal + K2mon. Çm) where K1reccv is a proportional gain (moment of inertia deviation) determined from the vehicle specifications and K2recv is a gain that changes depending on the vehicle speed V. An example of the K2recv gain is shown in Fig. 11. As shown in Fig. 11, for example, the K2recv gain has a large value in a

Petition 870170030306, of 05/08/2017, p. 40/67

33/54 low speed region. When the vehicle speed V reaches a certain value, the K2recv gain becomes inversely proportional to the vehicle speed V. Then, when the vehicle speed V reaches another certain value, the K2recv gain becomes equal to a constant value that it's a small value. In addition, a 5 K1mon clamping gain is a value using vehicle speed V as a function. In addition, a K2mon clamping gain is a value using the vehicle speed and observation period for Tt as functions.

In the formula (16) mentioned above, the target deviation moment Ms increases when the deviation angle φ between the vehicle MM and the white line 200 and the deviation rate constantly generated by additional driver steering increases.

Alternatively, the target deviation moment Ms can be calculated from formula (17) described below. This formula (17) is synonymous with multiplication of formula (16) mentioned above by a K3 gain (= 1 / Tt ² ). The K3 gain (= 15 1 / Tt ² ) is such a gain that it decreases as the observation time period ahead Tt increases.

Ms = K1recv.AXb / (L.Tt ² ) ....... (17)

When the formula (17) mentioned above is used, the following is obtained. The shorter the time of observation forwards Tt is, the greater the amount of control. In other words, when the forward observation time period is set to be short, the amount of control at the start of control is large. Meanwhile, the forward observation time period is set to be long, the amount of control is small. As a result, making the amount of control corresponding to setting the observation time period to 25 Tt forward allows a driver to be provided with a control that adjusts the situation and has less of an awkwardness.

Note that the Fout_obst judgment described above is carried out to predict a future change of range in the information bases.

Petition 870170030306, of 05/08/2017, p. 41/67

34/54 <Step S120>

In step S120, the brake / steering force control unit 8 calculates a command to generate the target deviation moment Ms for obstacle avoidance, sends the calculated command, and then returns to the first process.

Here, in the first realization, descriptions are given below of an example of the case where the deviation moment is generated through the use of a brake / steering force as a meaning for generating Ms deviation moment to avoid the obstacle.

Note that when the steering reaction force control device is used as the means for generating deviation moment, the ontro10le unit of brake force / steering 8 can calculate a reactive steering force Frstr as the command for generate the target deviation moment Ms, where Frstr = Ka.Ms, and then send the reactive steering force Frstr to the reactive steering force device to generate the reactive force. Note that, Ka mentioned earlier is a predetermined coefficient to convert the deviation moment to reactive steering force, which is obtained from an experiment or similar.

In addition, when a steering angle control device is used as the means for generating the offset moment, the brake / steering force control unit 8 can calculate a steering angle STR0 as the command to generate the steering moment. target deviation Ms, where STR0 = Kb.Ms, and then send the steer angle STR0 to the steer angle control device to control the steer angle. Note that, Kb mentioned above is a predetermined coefficient to convert the deviation moment to the steering angle, which is obtained from an experiment or similar.

In addition, when a steering force control device such as an energy steering system is used as the means for generating deviation moment, the brake / steering force control unit 8 can calculate the steering force (steering torque) of the steering force control device as the command to generate the target deviation moment Ms, where STR

Petition 870170030306, of 05/08/2017, p. 42/67

35/54 trg = Kc.Ms, and then send the steering force to the steering force control device to control the steering force. Note that, Kc mentioned earlier is a predetermined coefficient to convert the deviation moment to the steering force, which is obtained from an experiment or similar.

Additionally, when the generation of a brake force between the right and left wheels of the vehicle is used as the means of generating the offset moment, the brake force / steering control unit 8 calculates the command to generate the offset moment. target Ms as described below.

When the target deviation moment Ms is zero, that is, when a judgment results from such a condition that the deviation moment control is not to be performed is obtained, the next control is performed. As shown in formula (18) and formula (19) described below, hydraulic brake pressures (hydraulic brake pressures) Pmf, Pmr are sent to the brake fluid pressure control unit 7 as hydraulic brake pressures Psi targets ( i = fl, fr, rl, rr) of the respective wheels.

Then, the brake fluid pressure control unit 7 controls the fluid pressure circuit, and the hydraulic brake pressures of the respective wheels are thus controlled to be equal to the target hydraulic brake pressures Psi (i = fl, fr, rl, rr).

Psfl = Psfr = Pmf ....... (18)

Psrl = Psrr = Pmr ...... (19) where Pmf is a front wheel brake hydraulic pressure and Pmr is a rear wheel brake hydraulic pressure. The rear wheel brake hydraulic pressure Pmr has a value calculated based on the rear wheel brake hydraulic pressure Pmf in consideration of the front / rear distribution. For example, when the driver performs a braking operation, the hydraulic Pmf brake pressures,

Pmr have respective values corresponding to an operating amount (hydraulic pressure of master cylinder Pm) of the brake operation.

Meanwhile, when the absolute value of the target deviation moment Ms is greater than zero, that is, when a control

Petition 870170030306, of 05/08/2017, p. 43/67

36/54 obstacle avoidance is achieved, the following process is carried out.

Specifically, a target front wheel hydraulic pressure difference APsf and a target rear wheel hydraulic pressure difference APsr are calculated based on the target moment deviation Ms. To be more specific, the target brake hydraulic pressure differences APsf, APsr are calculated respectively from formula (20) and formula (21) described below.

APsf = 2.Kbf. (Ms.FReason) / Tr ....... (20)

APsr = 2.KBr. (Msx (1-FR reason) / Tr ....... (21) where FR ratio is a limit for fixation, Tr is a tread, Kbf and Kbr are conversion factors for the respective front and rear tires, which are used to convert braking forces to hydraulic brake pressures.

Note that the tread tr described above is the same for the front and rear for convenience. In addition, Kbf, Kbr are factors determined in advance from brake specifications.

As described above, the brake forces generated on the wheels are distributed according to the size of the target deviation moment Ms. In other words, the hydraulic pressure differences of the target brake APsf, APsr are each provided with a predetermined value, and thus a difference in brake force is generated between the left and right front wheels and between the left and right rear wheels. The differential hydraulic brake pressure differences calculated APsf, APsr are used to calculate the possible hydraulic brake pressures target Psi (i = fl, fr, rl, rr) of the respective wheels.

Specifically, when the direction of execution of the Dout_obst control is left, that is, when the obstacle avoidance control against the SM obstacle on the left side is to be performed, the hydraulic brake pressures target Psi (i = fl, fr, rl, rr) of the respective wheels are calculated from the formula (22) described below.

Psfl = Pmf,

Psfr = Pmf + APsf,

Petition 870170030306, of 05/08/2017, p. 44/67

37/54

Psrl = Pmr,

Psrr = Pmr + APsr ....... (22)

Meanwhile, when the direction of execution of the Dout_obst control is right, that is, when the obstacle avoidance control against the SM obstacle on the right side is to be carried out, the hydraulic brake pressures target Psi (i = fl, fr , rl, RR) of the respective wheels are calculated from the formula (23) described below.

Psfl = Pmf + APsf,

Psfr = Pmf,

Psrl = Pmr + APsr,

Psrr = Pmr ....... (23)

In the formula (22) and formula (23) mentioned above, the difference in brake force between the left and right wheels is generated in a way that the braking forces on the wheels on the obstacle avoidance side are fixed to be greater.

In addition, as shown in formula (22) and formula (23), the target brake hydraulic pressures Psi (i = fl, fr, rl, rr) of the respective wheels are calculated in consideration of the brake operation by the driver, Istoé, the hydraulic brake pressures Pmf, Pmr.

Then, the brake force / steering control unit 8 sends the target hydraulic brake pressures thus calculated Psi (i = fl, fr, rl, rr) of the respective wheels as the brake fluid pressure command values to the brake fluid pressure control unit 7. Then, the brake fluid pressure control unit 7 controls the fluid pressure circuit, and the brake fluid pressures of the respective wheels are controlled to be equal to the hydraulic pressures. brake Psi targets (i = fl, fr, rl, rr).

(Operation)

In the following, an operation of the first embodiment is described.

Fig. 12 is a view to explain the operation of the first embodiment.

Petition 870170030306, of 05/08/2017, p. 45/67

38/54

It is assumed that the MM vehicle is currently traveling in the center of the vehicle travel range. In this case, obstacle avoidance control determination flag Fout_obst = off and Freturn return flag = 0. In addition, the obstacle distance X2obst being the determination limit for the start of control of MM vehicle prevention from approaching SM side obstacle distance is X2obst_0, which is the obstacle distance in the normal state.

It is assumed that, from this state, the MM vehicle moves in the direction of SM obstacle (adjacent vehicle). At this time, the predicted vehicle position AXb is first calculated as a distance between the current lateral position of the 10 MM vehicle and the lateral position of the MM vehicle after the forward observation time period Tt, on the basis of the φfront offset angle, the angular deviation speed φπ, and the like that are the travel status of the MM vehicle (step S70 of Fig. 3).

Then, when AXb> X2obst (= X2obst_0) + X0 is set at a time tl of Fig. 12 (see, Fig. 4 also), the vehicle steering support control to avoid the obstacle is determined to be initiated, and obstacle avoidance control flag Fout_obst = on is set (step S90). As described above, the start of control is determined when the predicted vehicle position AXb reaches the side position of the bandwidth direction (X2obst + X0), the predicted vehicle position AXb being the future vehicle position at the time when the vehicle adjacent 20 SM is detected. In other words, control is initiated when the lateral position of the MM vehicle after the forward observation time period Tt reaches the control start position 60 being an off-line position 200 through X2obst in the direction of bandwidth .

When the start of vehicle steering support control is determined, the target deviation moment Ms is calculated as the control quantity on the basis of the predicted vehicle position AXb (step S110), and the brake / steering difference between the left wheels and the vehicle's right is controlled so that the target deviation moment Ms is generated (step S120). Specifically, In Fig. 12, the braking forces

Petition 870170030306, of 05/08/2017, p. 46/67

39/54 on the left wheels of the vehicle are controlled to be larger than those of the right wheels at the bases of the target deviation moment Ms. This generates a deviation moment in one direction such that the vehicle is avoided from approaching the adjacent vehicle SM (in a direction from the center of the vehicle travel range), and the 5 MM vehicle is thus controlled.

When the driver recognizes the SM side obstacle due to this vehicle steering support control to avoid the obstacle and the MM vehicle attempts to return to the center of the lane, on the return course, the predicted vehicle position AXb returns into the interior of the lane. control start position 60 at time t2. In other words, AXb <X2obst + X0 is established. This causes the Fout_obst + obstacle avoidance control flag to be set.

At this time, the previous value of the Fout_obst obstacle avoidance control flag is set. Thus, the return flag Freturn = 1 is fixed (step S80 of Fig. 5). Then, the start control position 60 fixed in an off-line position 200 in the direction of bandwidth by the predetermined reference obstacle distance X2obst_0 at this time is still fixed out in the direction of bandwidth by the amount of correction õyc . In other words, when X2obst = X2obst_h (step S806) is fixed, the corrected control start position 60 is located closer to the obstacle than the control start position 20 60 before correction, that is, further away from the trip. Thus, the frequency of the start of control is reduced.

Then, when the predetermined time period elapses from time t2 in which the return flag Freturn = 1 is set (S809 in Fig. 5), the return flag Freturn = 0 is set (S810 in Fig. 5) and the obstacle distance X2obst is retor25 nothing for the reference value (X2obste_0) (S808 of Fig. 5) at a time t3.

In contrast to the first embodiment in which the control start position 60 is corrected, the control start position 60 is not corrected in the conventional process. In the conventional process, the control start position 60 does not change even

Petition 870170030306, of 05/08/2017, p. 47/67

40/54 when the predicted vehicle position AXb returns into the control start position 60 at a time 12 after the control is started at a time t11 of Fig. 13. In other words, the obstacle distance X2obst is maintained at reference obstacle distance X2obst_0.

In the same way, when the driver's steering causes the predicted vehicle position AXb to be unstable after time tI2 and AXb> Xobst (= X2obst_0) + X0is therefore established in time t13, the control is started again at this same point . Specifically, for example, when the travel direction of the MM vehicle is an inland direction at a time point between time t12 and 10 time t13, the driver sometimes performs a steer operation out of the range ( ascending in the drawing) to adjust the travel direction of the MM vehicle to the direction extending from the lane (to reduce the forward angle ^ of the MM vehicle). In such a case, the driver performs the steering operation to change the position (angle of deviation) of the MM vehicle in the process of returning 15 MM vehicle to the interior of the lane. The control intervenes again even in a situation where the driver is performing a steering that has no intention of moving the vehicle in the direction of an obstacle.

As described above, there is a case where the obstacle avoidance control intervenes again although the driver recognizes the Sm obstacle and is trying to return to the inside of the travel lane. This control gives the driver the feeling of strangeness. In particular, the longer the forward observation time Tt, the more likely the future vehicle position after the fixing time period is unstable through driver steering. Likewise, in the process of determining the start of obstacle avoidance control 25 through the use of future vehicle position, a new unnecessary control is likely to be performed, and the driver is more likely to receive the feeling of awkwardness.

On the other hand, in the first realization, when the future vehicle position 150

Petition 870170030306, of 05/08/2017, p. 48/67

41/54 returns to the inside of the control start position 60 after the control start (time t1), the control start position 60 is changed to a more distant position from the travel range by the amount of correction õyc to the predetermined period of time (period from time t2 to time t3 of Fig. 12) after the return. In the same way, even when the vehicle forecast AXb is unstable and AXb> X2obst_0 + X0 is therefore established in the period (predetermined time period) from time t2 to time t3, the start of control is suppressed. This is because the distance from the white line 200 to the control start position 60 is changed from the reference obstacle distance X2obst_0 to the corrected obstacle distance X2obst_h and AXb <X2obst_h + X0 is as established. As a result, the obstacle avoidance control is rendered inactive. Therefore, the feeling of awkwardness provided to the driver due to unnecessary control intervention can be reduced.

In addition, the amount of suppression of the control start determination can be adjusted by fixing the corrected obstacle distance X2obst_h according to the lane width of the vehicle travel lane, the distance to the opposite side obstacle on the opposite side to the avoidance control target, white line recognition certainty, obstacle recognition certainty and the like. Specifically, the determination of start of control is largely suppressed in the following ways. The narrower (smaller) the bandwidth, or the shorter (shorter) the distance between the obstacle on the opposite side and the MM vehicle, the greater the amount of õyc correction fixed (see Fig. 6). In addition, the lower the certainty of white line recognition, the greater the amount of õyc correction set (see Fig. 7). In addition, the lower the certainty of obstacle recognition, the greater the amount of õyc correction set (see fig. 8). In addition, the lower the stability of the vehicle position of the MM vehicle, the smaller it is (the greater the amount of movement change of the MM vehicle), the greater the amount of fixed correction õyc (see Fig. 9). Therefore, the determination of start of control

Petition 870170030306, of 05/08/2017, p. 49/67

42/54 vehicle steering support can be made while obtaining a reduction in the feeling of awkwardness provided to the driver.

Note that, in Fig. 1, the radar devices 24L / 24R form the side obstacle detection unit 25. In addition, step S70 in Fig. 3 corresponds to the future position predictor 8A, step S80 corresponds to the suppression part of control 8Ba, step S90 corresponds to the control start determination part 8B, and steps S100 to S120 correspond to vehicle controller 8C.

(Effects) (1) Control start determination part 8B determines to start control when the lateral position of the MM vehicle in the bandwidth direction reaches the control start position 60 with the predetermined lateral position in the width direction. track. When the control start determination part 8B determines the control start, vehicle controller 8C controls the MM vehicle causing a deviation moment towards the center of the vehicle travel range to be applied to the MM vehicle.

When the lateral position of the MM vehicle in the bandwidth direction moves from a position outside the control start position 60 to a position approaching the vehicle travel range within the control start position 60 in the bandwidth direction , the control suppression part 8Ba suppresses the deviation moment application control for the MM vehicle until a predetermined control state retention period elapses after the lateral position of the MM vehicle in the direction of the bandwidth moves to the position approaching the vehicle travel range, compared to the period before moving to the position approaching the vehicle travel range.

As described above, when the lateral position of the MM vehicle in the bandwidth direction changes from the outside to the inside of the control start position 60, the MM vehicle control is suppressed by suppressing control start determination until that the state retention period has elapsed

Petition 870170030306, of 05/08/2017, p. 50/67

43/54 control. In the same way, even when the lateral position of the MM vehicle in the direction of bandwidth is unstable, unnecessary control intervention is suppressed.

This allows the vehicle steering support to be properly carried out while suppressing the start of control which creates the feeling of awkwardness for the driver.

(2) The future position predictor 8A predicts the future vehicle position 150 (predicted vehicle position AXb being the distance from the vehicle's current lateral position to the future vehicle position 150) after the fixed period of time (period observation time mode forward Tt) from the current time.

In the same way, cancellation of unnecessary control and intervention of a desired control to be performed can be easily achieved.

(3) When the lateral position of the MM vehicle in the bandwidth direction moves from a position outside the control start position 60 in the bandwidth direction to a position approaching the vehicle travel range within the start position of control 60 in the bandwidth direction, the control suppression part 8Ba suppresses the control by changing the position of control start 60 to a position further away from the vehicle travel range in the bandwidth direction.

This allows the determination of control start to be suppressed in a relatively simple way.

(4) A bandwidth detector (image capture unit 13) detects the line width of the vehicle travel range. The control suppression part 8Ba suppresses more control as the bandwidth of the vehicle travel range detected by the bandwidth detector becomes narrower (image capture unit 13) (see Fig. 6).

As described above, the narrower the vehicle travel bandwidth, the greater the amount of change (amount of correction õyc) of the position

Petition 870170030306, of 05/08/2017, p. 51/67

44/54 control start 60 fixed so that the control start position 60 is furthest from the vehicle travel range (closest to SM obstacle). Therefore, the following can be said. The narrower the lane width, the more frequently the driver performs the steering operation. Thus, the unnecessary control intervention 5 can be suppressed in a situation where the bandwidth is narrow and the position of the MM vehicle is likely to overturn.

(5) The obstacle distance detector (24L / 24R radar devices) detects the distance between the MM vehicle and the opposite side obstacle on one side for which the MM vehicle is to move laterally by controlling the controller of vehicle 8C. The control suppression part 8Ba further suppresses the control as the distance between the MM vehicle and the opposite side obstacle detected by the obstacle distance detector (24L / 24R radar devices) becomes smaller (see Fig. 6) .

As described above, the shorter the distance to the opposite side obstacle on the opposite side for target obstacle control SM, the greater the amount of change (correction amount õyc) from the control start position 60 fixed in a way such that the control start position 60 is furthest from the vehicle travel range (closest to the SM obstacle). Therefore, the following can be said. The shorter the distance to the opposite side obstacle on the opposite side to the target SM control obstacle, the more frequently the driver performs the steering operation. Thus, unnecessary control intervention can be suppressed in a situation where the distance to the opposite side obstacle on the opposite side to the target SM control obstacle is short and the vehicle's position is likely to overturn.

(6) The vehicle posture detector (steering angle sensor 19, deviation rate sensor, acceleration sensor, lateral force sensor, and the like) detects the vehicle posture stability of the MM vehicle. The control suppression part 8Ba further suppresses control when the vehicle's posture stability

Petition 870170030306, of 05/08/2017, p. 52/67

45/54

MM detected by the vehicle posture detector (steering angle sensor 19, deflection rate sensor, acceleration sensor, lateral force sensor, and the like) becomes smaller (when the amount of vehicle movement change MM becomes greater (see Fig. 9).

This allows unnecessary control intervention to be suppressed in a situation where the change in movement of the MM vehicle is likely to change to a relatively large degree, such as a situation where the MM vehicle is traveling through a curve.

(7) A lane dividing line detector (image capture unit 130 detects a lane dividing line (white line 200) from the vehicle travel lane. With lane dividing line detection, the lane dividing detector range (image capture unit 13) determines the certainty of recognition of the range dividing line.The start control position 60 is fixed in a position that is away from the position of the range dividing line by the predetermined distance X2obst in the bandwidth direction, the strip dividing line detected by the strip dividing line detector The control suppression part 8Ba further suppresses control as the certainty of recognizing the strip dividing line determined by the strip dividing line detector becomes narrower (see Fig. 7).

In other words, the control start position 60 is fixed at the position that is away from the lane dividing line (white line 200) by the predetermined distance X2obst, and the control start is determined based on whether the lateral steering distance AX2 from the white line 200 to an estimated future position (vehicle future position 150) reaches the limit X2obst.

In addition, the greater the variation (variation in the lateral position) in the white line recognition result 200 (range dividing line), the greater the amount of change (correction amount õyc) from the control start position 60 fixed so that the control start position 60 is furthest from the vehicle travel range (closest to SM obstacle). Therefore, the following can be said.

Petition 870170030306, of 05/08/2017, p. 53/67

46/54

When the certainty of recognition of the range dividing line is low, the lateral position of the recognized range dividing line varies and so does the start control position 60 variation. Therefore, unnecessary control intervention is more likely to occur . Therefore, the less certainty of recognition of the lane dividing line, the more control is suppressed. This allows unnecessary control intervention to be suppressed even when the lateral position of the recognized lane dividing line varies.

(8) The side obstacle detection unit 25 detects the existing SM side obstacle for the MM vehicle side. The inspection start determination part

8B determines the control start when the lateral position of the MM vehicle in the bandwidth direction reaches the control start position 60 in the state where the side obstacle detection unit 25 is detecting the side obstacle SM.

This allows the vehicle steering support control against the SM side obstacle to be properly carried out while suppressing the control start 15 which causes the driver to feel awkward.

(9) With SM side obstacle detection, the side obstacle detection unit 25 determines the certainty of side obstacle recognition. The control start position 60 is fixed based on the MM vehicle and the distance (relative distance in the lateral direction AO) (between the MM vehicle and the SM obstacle) in the bandwidth direction that is detected by the lateral obstacle detection unit 25. The control suppression part 8Ba further suppresses the control when the certainty of lateral obstacle recognition determined by the side obstacle detection unit 25 becomes smaller (see Fig. 8).

In other words, the start of control is determined based on whether the predicted future position AXb of the MM vehicle reaches the relative lateral direction distance AO between the MM vehicle and the detected obstacle SM.

In addition, the greater the variation in the SM obstacle recognition result, the greater the amount of change (correction amount õyc) of the position

Petition 870170030306, of 05/08/2017, p. 54/67

47/54 control start 60 fixed so that the control start position 60 is furthest from the vehicle's travel range (closest to the SM obstacle). This allows the unnecessary control intervention to be suppressed even when the SM obstacle swings or even when a balance ratio between the MM vehicle and the SM obstacle is relatively large.

(10) The start of control is determined when the lateral position of the MM vehicle in the direction of the bandwidth reaches the start position of control 60, the lateral position being predetermined in the direction of the bandwidth which is an approach prevention indicator for the MM vehicle, and the moment of deviation in the center direction of the vehicle travel range is applied to the MM vehicle to control the MM vehicle. In addition, when the lateral position in the AXb bandwidth direction of the MM vehicle moves from an out of control start 60 position in the travel bandwidth direction to a position approaching the travel band within control start position 60 in the bandwidth direction, the deviation moment application control for the MM vehicle is suppressed until the predetermined control state retention period elapses after the bandwidth direction side position AXb of the MM vehicle moves to the position approaching the vehicle travel range, compared to the period before moving to the position approaching the vehicle travel range.

This allows the approach to the SM obstacle to be appropriately avoided while suppressing the start of control which renders the driver feeling awkward.

(Second Realization)

In the following, a second embodiment of the invention is described.

In the first realization described above, the control start is suppressed by adjusting the control start position 60 used for the control start. Meanwhile, in the second realization, the start of control is suppressed by adjusting an observation time period forward Tt.

Petition 870170030306, of 05/08/2017, p. 55/67

48/54 (Configuration)

A basic configuration of the second embodiment is the same as that of the first embodiment described above.

Fig. 14 is a flowchart showing an obstacle avoidance control processing procedure performed by a brake / steering force control unit 8 in the second embodiment.

In obstacle avoidance control processing, the same processes as those in Fig. 3 obstacle avoidance control processing are performed, except that a step S65 process is added and a step S80 process is different from that of the obstacle avoidance control processing shown in Fig. 3. Likewise, portions of the processing that are different are mainly described below.

<Step S65>

In step S65, the steering / brake control unit 8 adjusts the period of observation time forward Tt 15.

In step S65, when a Fout_obst obstacle avoidance control flag changes from on to off, the forward observation time period Tt fixed in step S60 mentioned above is adjusted based on the following formula

Tt <- Tt.Kt ....... (24) where Kt is a gain and Kt <1. As similar to fixing the correction amount õyc in the first realization described above, the gain Kt is fixed according to a lane width of a vehicle travel lane, a white line recognition certainty, an obstacle recognition certainty, and the like, in a manner such that the more likely the vehicle position is to swing close to a used determination position for the start of control, the shorter the observation time for fixed Tt is shorter.

In addition, the Kt gain can be fixed so that the longer the pe

Petition 870170030306, of 05/08/2017, p. 56/67

49/54 time period elapsing after the obstacle avoidance control flag Fout_obst changes from on to off, closer to “1” the Kt gain is fixed.

<Step S80>

In step S80, the brake force / steering control unit 8 fixes a control start position 60 based on a predetermined obstacle distance X2obst (that is, the reference obstacle distance X2obst_0 at the first realization). Note that a relative AO lateral steering distance between an MM vehicle and an SM obstacle can be used as the X2obst obstacle distance.

(Operation)

In the following, an operation of the second embodiment is described.

It is assumed that a vehicle position (forward observation point 150) after the forward observation time period Tt currently returns to the inner lane direction of the start control position 60 after the start of the avoidance control obstacle. In this case, the Fout_obst obstacle avoidance control flag changes from on to off.

First, the forward observation time period Tt is fixed which is used to calculate a future position of vehicle AXb being a distance between the current lateral position of the vehicle and the future lateral position (observation point for 20 forward 150) of the vehicle. Here, the forward observation time period Tt is a shorter period of time than that in a normal state (step S65). Then, the predicted position of vehicle AXb is calculated as a distance between the current lateral position of the vehicle and the lateral position of the vehicle after the forward observation time period Tt, at the bases of an angle of forward deviation, an angle speed deviation Am, and similar which are the travel states of the MM vehicle (step S70).

Then, the vehicle steering support control to avoid the obstacle is initiated when a lateral steering distance AX2 being a value obtained through

Petition 870170030306, of 05/08/2017, p. 57/67

50/54 subtracting a current lateral displacement X0 from the predicted vehicle position AXb on a side where the obstacle SM (adjacent vehicle) is detected reaches or exceeds the predetermined obstacle distance X2obst (when the lateral position of the vehicle MM after the period of forward observation time Tt, that is, the forward observation point 150 is out of control start position 60 in the bandwidth direction) (step S90). Here, since the forward observation time period Tt is adjusted to be shorter, the forward observation point 150 becomes closer to the current position of the MM vehicle. In the same way, the control start is suppressed (the frequency of the control start is reduced) compared to a period before correction of the forward observation time period Tt.

Likewise, even when the predicted vehicle position AXb is unstable for a predetermined period of time after the vehicle position after the forward observation time period Tt returns to the inside of the start control position 60 in the wide direction on the lane, unnecessary control is suppressed by new intervention and the feeling of strangeness provided to the driver can be reduced.

In addition, the reduction in the observation time ahead Tt makes the predicted position of the AXb vehicle smaller. Likewise, when a target deviation moment Ms is calculated from the formula (16) mentioned above, a control quantity (target deviation moment Ms) is small even if control is initiated. This leads to the suppression of change in vehicle behavior when control intervenes.

Meanwhile, when the target deviation moment Ms is calculated from the formula (17) mentioned above, the amount of control increases even if the predicted position of vehicle AXb decreases. Thus, a control that adjusts the situation independently of the setting of the observation time period ahead Tt and that does not provide a feeling of awkwardness for the driver can be obtained.

Note that step S65 of Fig. 14 corresponds to the DePetition suppression part 870170030306, of 05/08/2017, p. 58/67

51/54 start termination.

(Effect) (11) A control suppression part 8Ba suppresses the determination of control start by reducing fixation time (forward observation time period Tt) used when a future position predictor 8A predicts the future position vehicle (expected vehicle position AXb).

This allows the determination of control start to be suppressed in a relatively simple way.

(Modified Example) (1) In each of the previously mentioned embodiments 1 and 2, descriptions are given of the case where the invention is applied to the obstacle avoidance control to avoid contact between the MM vehicle and the Sm side obstacle, such as the vehicle steering support control. However, the intervention is also applicable to a lane departure prevention control where the MM vehicle is controlled 15 using the lane as a target, regardless of the existence of the SM side obstacle.

Specifically, the invention is applicable to a lane departure prevention control that prevents the MM vehicle from leaving the lane by applying a moment of deviation to it, regardless of the existence of the SM lateral obstacle, when the lateral position (starting point) forward observation 150) from the vehicle position after the forward observation time period20 reaches control start position 60 or moves out of control start position 60 in the direction of bandwidth.

In this case too, the departure of the lane can be appropriately prevented while the start of control that provides the sensation of strangeness for the motorist is suppressed.

(2) In each of the achievements 1 and 2 mentioned above, the descriptions are given of the case where the determination of start / cancellation of control is made based on the future position 150 of the MM vehicle after the predetermined time (pePetition 870170030306, from 08 / 05/2017, page 59/67

52/54 forward observation time period Tt) (ie based on the predicted position of vehicle AXb). However, the determination of start / cancellation of control can also be made simply based on the lateral position (lateral displacement X0) of the MM vehicle. In other words, the forward observation time period Tt 5 can be set to zero.

(3) In each of the achievements 1 and 2 mentioned above, descriptions are given of the case where the determination of start of control is suppressed for the predetermined period of time (a certain period of time) after the vehicle position returns to the position control start 60, such as the control state retention time period. However, the control start determination can be suppressed until the MM vehicle travels a predetermined distance (up to the time required for the MM vehicle to travel the predetermined distance), after the vehicle position returns to the control start position 60 .

In addition, the control start determination can be suppressed until the vehicle position reaches a suppression cancel position for the control start determination suppression cancellation (until the time required for the MM vehicle to reach the control position has elapsed. cancellation cancellation), after the vehicle position returns to the control start position 60. The cancellation cancellation position is fixed, for example, in the center of the vehicle's travel range.

In other words, the start of control simply needs to be suppressed for a predetermined period of time such as until a predetermined period of time has elapsed, until the MM vehicle travels the predetermined distance, or until the vehicle position reaches the cancel position. of suppression, after the vehicle position 25 returns to the control start position 60.

(4) In each of the achievements 1 and 2 mentioned above, the period in which the start of control is suppressed can be fixed in such a way that the more likely the vehicle position is to swing close to the determination limit for the beginning

Petition 870170030306, of 05/08/2017, p. 60/67

53/54 of control, the longer the fixed period. Thus, the control of vehicle steering support with the sensation of strangeness still reduced can be performed.

(5) In each of the aforementioned achievements 1 and 2, the determination limit (or the forward observation time period Tt) such as the obstacle distance X2obst can gradually return to its normal value when the suppression of control start determination is canceled. This allows for a smooth transition from the state where the control start determination is suppressed to the normal control state.

(6) In each of the achievements 1 and 2 mentioned above, the start control chromatography is adjusted by adjusting the forward observation time period Tt or the determination limit such as the obstacle distance X2obst. Instead, the timing of the start of the control can be delayed by multiplying an adjustment gain (<1) for the predicted position of vehicle AXb calculated in step S70. In this case too, effects similar to those of achievements 1 and 2 mentioned above are obtained.

In addition, the control start timing can be adjusted by multiplying the adjustment gain to the predicted vehicle position AXb in the condition used to determine the control start timing in step S90. In this case, even when the timing of the start of the control is adjusted through the use of adjustment gain, the amount of control (target deviation moment Ms) in the control operation is not affected by the adjustment gain.

Whole JP N patent application content ^o · 2009-177422 (filing date in Japan: July 30, 2009) and Japanese patent application No. 2010-133851 · (filing date in Japan: June 11, 2010 ) which are basic patent applications filed in Japan are hereby incorporated, and are protected from erroneous translations and omissions.

The content of the invention has been described using first and second embodiments and their modified examples. However, it is obvious to those

Petition 870170030306, of 05/08/2017, p. 61/67

54/54 technique that the invention is not limited to these descriptions and several modifications and improvements can be made to it.

Industrial Applicability

In the invention, when the lateral position of the vehicle (MM) in the bandwidth direction moves into the control start position (60) after the control start determination is made, the control is suppressed for the predetermined period after the lateral position move inward. In the same way, even when the vehicle's lateral position (MM) in the direction of lane width is unstable, control can be suppressed. As a result, the feeling of awkwardness provided to the driver can be reduced.

Claims (11)

1. Vehicle steering support device, comprising: a control start determination part (8B) configured to determine a control start when a vehicle side position (MM) in 5 a bandwidth direction reaches a control start position (60) with a predetermined lateral position in the bandwidth direction; a vehicle controller (8C) configured to control the vehicle (MM) by applying a moment of deviation in the center direction of a vehicle trip range for the vehicle (MM) when the control start part of 10 control (8B) determines the start of control; in which vehicle steering support device is CHARACTERIZED for still understanding: a control suppression part (8Ba) configured for, when the vehicle's lateral position (MM) in the bandwidth direction moves from a position 15 out of the control start position (60) in the bandwidth direction to a position approaching the vehicle travel range within the control start position (60) in the direction of the lane width, suppress the determination of control start of application of the moment of deviation to the vehicle (MM) for a predetermined period after the lateral position of the vehicle (MM) in the direction of bandwidth move to the position approaching the vehicle travel range, compared to a period before moving to the position approaching the vehicle travel range. 2. Vehicle steering support device, according to claim 1, CHARACTERIZED by further comprising a future position forecasting part (8A) configured to predict a future position (150) of the vehicle (MM) after a 25 fixing time period (Tt) from a current time, where the lateral position of the vehicle (MM) in the bandwidth direction is the future position (150) predicted by the future position forecasting part (8A). 3. Vehicle steering support device according to claim 1 or 2, CHARACTERIZED by the fact that the control suppression part Petition 870170030306, of 05/08/2017, p. 63/67 2/4 (8Ba) suppresses the determination of the start of control by changing the position of the start of control (60) to a position further away from the vehicle travel range in the direction of lane when the lateral position of the vehicle (MM ) in the direction of bandwidth moves from position out of control start position (60) 5 to the position approaching the vehicle travel range within position of control start (60) in bandwidth direction . 4. Vehicle steering support device, according to claim 2, CHARACTERIZED by the fact that the control suppression part (8Ba) suppresses the control by reducing the clamping time period (Tt) used in the 10 forecasting the future position (150) of the vehicle (MM) through the future position forecasting part (8A). 5. Vehicle steering support device according to any one of claims 1 to 4, CHARACTERIZED by further comprising a lane width detector (13) configured to detect a lane width of the road 15 of the vehicle, in which the control suppression part (8Ba) further suppresses the control as the bandwidth of the vehicle travel range detected by the bandwidth detector (13) becomes narrower. 6. Vehicle steering support device, according to any 20 of claims 1 to 5, CHARACTERIZED by further comprising an obstacle distance detector (24L, 24R) configured to detect a distance (AO) between the vehicle (MM) and an obstacle (SM) existing on one side for which the vehicle (MM) is to move laterally through the vehicle controller control (8C), in which the control suppression part (8Ba) further suppresses the control according to the distance (AO) between the vehicle (MM) and the obstacle (SM) detected by the obstacle distance detector (24L, 24R) becomes smaller. Vehicle steering support device according to any one of claims 1 to 6, CHARACTERIZED by further comprising a vehicle posture detector (19) configured to detect the stability of a vehicle posture Petition 870170030306, of 05/08/2017, p. 64/67 3/4 vehicle vehicle (MM), where the control suppression part ( 8Ba) suppresses the determination of start of control more as the stability of the vehicle's vehicle posture (MM) detected by the vehicle posture detector (19) becomes less. 8. Vehicle steering support device according to any one of claims 1 to 7, characterized by further comprising a lane dividing line detector (13) configured to detect a lane dividing line (200) from vehicle travel lane, where the lane dividing line detector (13) determines a certainty of 10 recognition of the dividing line (200) when detecting the dividing line (200), the control start position (60) is fixed as a position away from a dividing line (200) position detected by the strip dividing line detector (13) by a predetermined distance (X2obst) in the direction of strip width 15, and the control suppression part (8Ba) further suppresses the control according to the certainty of dividing line recognition range (200) determined by the range dividing line detector (13) becomes smaller. 9. Vehicle steering support device, according to any 20 of claims 1 to 7, CHARACTERIZED by further comprising a lateral obstacle detector (25) configured to detect a lateral obstacle (SM) on the side of the vehicle (MM), wherein the control start determination part (8B) determines start of control when the lateral position of the vehicle (MM) in the direction of bandwidth reaches the start of control position (60) in a state where the side obstacle (SM) is detected by the side obstacle detector (25 ). 10. Vehicle steering support device according to claim 9, CHARACTERIZED by the fact that the side obstacle detector (25) determines a recognition certainty Petition 870170030306, of 05/08/2017, p. 65/67 4/4 of the side obstacle (SM) when detecting the side obstacle (SM), the control start position (60) is fixed based on a distance (AO) between the vehicle (MM) and the detected side obstacle (SM) by the side obstacle detector (25) in the bandwidth direction, and the control suppression part (8Ba) further suppresses the control as the lateral obstacle (SM) recognition certainty determined by the side obstacle detector (25) makes it become smaller. 11. A vehicle's steering support process, comprising: a control start determination step that determines the start of a control when a lateral position of a vehicle (MM) in a lane direction reaches a control start position (60) being a predetermined lateral position in the direction of bandwidth; a vehicle control step that controls the vehicle (MM) by applying a detour moment in the center direction of a vehicle to vehicle (MM) travel strip; where the steering support process of a vehicle is CHARACTERIZED because it still comprises: a control suppression step which, when the vehicle's lateral position (MM) in the lane width direction moves from a position outside the start control position (60) in the lane width direction to a position approaching the lane vehicle travel position within the control start position (60) in the direction of the lane width, suppresses the determination of control start of the application of the deviation moment for the vehicle (MM) for a predetermined period after the lateral position of the vehicle (MM) in the direction of bandwidth move to the position approaching the vehicle travel range, compared to a period before moving to the vehicle travel range approach position.